Selected Current Research
Asymmetries in Spoken Word Perception and Sound Change
My dissertation research concerns the connection between asymmetries in spoken word perception and asymmetries in rates of sound change in words of different frequencies. It combines a corpus-based case-study in New Zealand English with computational modeling, drawing heavily on results from psycholinguistic experiments in the literature.
Word Frequency Effects in Sound Change as a Consequence of Perceptual Asymmetries
We have built a computational exemplar-theoretic model for the perception/production loop to explore how items of different lexical frequencies may respond to changes in discriminability in the phonological space during sound change. The model predicts high-frequency words to change at a faster rate than low-frequency words when the change decreases discriminability, at a slower rate than low-frequency words when the change increases discriminability, and at the same rate as low-frequency words when the change does not affect discriminability. This paper forms one of the core chapters of my dissertation.
Something from nothing: pragmatic parsing of partitive possessives
I ran an experiment to test a Bayesian pragmatic model of parsing which predicts that the failure to clearly indicate an intended parse of a partitive plural possessive, via the phonological suppression of the possessive morpheme, inspires a meaningful inference when the intended parse is otherwise reasonably ambiguous. This psycholinguistic project brings together morphophonology, syntactic parsing, and the division of pragmatic labor.
Avoiding interpretation ambiguities with phonetic cues
I developed a Bayesian metric of the ease with which a listener of a dative alternation would correctly assign the thematic roles of recipient and theme if the preposition "to" were not heard, which I dubbed interpretability. I used this metric in a corpus study to test an information-based hypothesis relating to audience design, stating that, if speakers are aware of the information needs of their listener, they should strengthen the acoustic cues to /t/ in the word "to" in low-interpretability prepositional datives (when the listener would be likely to assign thematic roles incorrectly if the extra cue provided by "to" were not attended to).
Faithfulness and Exceptionality in Polish Stress
I describe new facts about the placement of secondary stress in inflected exceptional stems and compounds in Polish, which are incompatible with existing analyses based in monostratal Optimality Theory. I propose an alternative analysis based in a multistratal approach to OT which not only accounts for these new facts, but also efficiently harnesses the similarity in stress placement across different morphophonological domains.
Syntactic Embedding and Visbility of Morphophonological Structure
I present experiments that investigate gradience and variation in the propensity to suppress the English possessive morpheme /z/ with possessors ending in the plural morpheme /z/. I find that suppression becomes less necessary as the number of syntactic brackets between the two /z/s increases. I sketch formal implementations of this finding in representational and derivational morphophonological frameworks, and I argue that it implies that morphophonological processing must be guided by syntactic (or corresponding prosodic) structure in a stochastic fashion.
Attitudinal Dynamics in Language Change: Eh in NZ English.
I have conducted corpus-based exploration of variation and change in the use of the discourse particle eh in New Zealand. I find evidence that eh has spread from indigenous Māori to white Pākeha over time, and that this spread was facilitated by the decrease in social stigma attached to Māori. I link this result to psycholinguistic findings about the influence of social stigma on linguistic memory and suggest ways that it can be further tested experimentally.
Functional load and neutralization in sound change
[+] Early work presented at Language and Society, University of Auckland, 2012 [poster]
Two related projects come under this heading, both pertaining to the hypothesis that a phonological contrast will be more impervious to neutralization the more "work" it does in preventing ambiguity (its functional load). In the first project, I explore changes in the duration of long vowels in Māori, which are decreasing in all cases except /a:/. Using numerous corpus-based calculations suggested by the literature, I find that the /a:/~/a/ contrast carries much higher functional load than other vowel length contrasts, supporting the Functional Load Hypothesis. In the second (ongoing) project, I use the computational linguistic tool of word vectors to ask whether merger in varieties of English is less likely for contrasts that prevent a substantial amount of ambiguity in use, by having many minimal pairs whose members are used in similar lexical environments.